This evening same workunit has validation problem (inconclusive validation)
http://gene.disi.unitn.it/test/results.php?userid=42
What is the problem?
This is log:
<core_client_version>7.2.42</core_client_version>
<![CDATA[
<stderr_txt>
Start @ Thu Apr 17 13:49:26 2014
Start @ Thu Apr 17 19:47:34 2014
Start from checkpoint: 1
|11384|
Finish @ Thu Apr 17 20:23:18 2014
20:23:18 (2636): called boinc_finish
</stderr_txt>
]]>
____________
Reply
Quote
This evening same workunit has validation problem (inconclusive validation)
http://gene.disi.unitn.it/test/results.php?userid=42
It's user id or host id? Cannot open link
Host id 42 is, again, my work notebook with win 8.1 and 7.3.15 boinc client.
Now i downgrade to 7.2.42, latest stable....will see what happens
Reply
Quote
The host ID is #132 and the workunit are 7547 (task 15997) and 7499 (task 15900)
Reply
Quote
The host ID is #132 and the workunit are 7547 (task 15997) and 7499 (task 15900)
I have also some invalids...and some "validation inconclusive" (all on host id 42)
Reply
Quote
valtercProject administrator
Project tester
Send message
Joined: 30 Oct 13
Posts: 623
Credit: 34,677,535
RAC: 2
The validator logic is quite simple, a couple of output files is compared byte by byte and are marked valid if they are exactly the same. Both of the invalids of host #132 have a duplicate line in the stderr output, "Start @ ...", but stderr of Task 15994 (same host) has the same structure and has been marked valid. We need to investigate further.
Host #42 has a different problem. I just checked workunit 8063, it is inconclusive by now, which means that two results has been returned but are not equal (they have the same size but different content). This is what I get when trying to gunzip result #17041 (got from host #42)
This means that either the file was badly compressed client-side or there was some kind of transmission/communication error.
Reply
Quote